It isn't just the pattern of similarities that evidence common descent and evolution. The differences also provide us with big pieces of evidence.
The inspiration for this post comes from EvoGrad and Stephen Schaffner, from whom I will be borrowing various figures. Ebersberger et al. (2002) published on this subject, and could have served as direct or indirect inspiration for EvoGrad and Schaffner.
With credit given, let's move on to the evidence.
Mutations are heritable changes, specifically heritable changes in the DNA sequence of a genome. The type of mutation I will be discussing is a substitution mutation where one base is swapped out for a different base. For example:
AGGCTAATCG --original AGGGTAATCG --mutated
There are two main types of substitutions: transitions and transversions. They are called this because if the mutation is between two similar bases it is a transition, and a transversion if it is between two dissimilar bases. The two classes of bases are purines and pyrimidines. For clarity's sake, I like to refer to them as one ring and two ring bases, as shown in the picture below:
Due to the biochemistry of genetics, transitions tend to happen more often than transversions. That is, substitutions occur more often between bases that have the same number of rings. Even though there are two possible transversion mutations per base compared to just one possible transition, we still see more transitions than transversions.
Evograd compiled a total of 220,000 de novo (i.e. new mutations detected in experiments in green) human mutations from various papers and compared them to 78.6 million substitutions found in the existing human population (i.e. the standing variation in the human population in blue) from public databases. This is what that comparison looks like:
Like I stated earlier, transitions outnumber transversions in this figure. The first set of bars are the transitions, and the other three sets of bars are the transversions. Also, the rate at which these mutations occur in real time matches the standing variation in the human population. In other words, this is smoking gun evidence that the process we observe creating mutations in real time is responsible for the variation we see in the human population. The fingerprint produced by the natural process of mutation is measurable and present in the human population.
But what if we do the same thing for a comparison of the human and chimp genome? The model for common descent and evolution states that humans and chimps share a common ancestor. Therefore, this model predicts that our lineages started from the same ancestral genome and population. As our lineages diverged, the same process of mutation should have created differences between those lineages. Therefore, if this model is correct then we should see the same fingerprint when we compare the human and chimp genomes.
Wouldn't you know it, there's that fingerprint. In fact, let's extend it out to other primates:
There's that same fingerprint, just as we would expect from common ancestry and evolutionary mechanisms.
This is smoking gun evidence for common ancestry. This evidence is exactly what we would expect to see if our models are true.
In science we like to use statistics to measure the fit between data and model, so we should do the same for spectrum of mutations. Luckily, Francioli et al. (2015) have already done this for us. They compared mutations in the context of three base pair motifs, as well as in the context of CpG and non-CpG mutations (a subject I will probably touch on later).
Figure 6 | Correlation between observed de novo mutation rates and human/chimp substitution rates for mutation types in different trinucleotide contexts. De novo mutation rate spectrum (Y-axis) is plotted against substitution rate spectrum inferred from human vs chimp comparison (X-axis). Each dot represents a type of mutation in a specific trinucleotide context. The Pearson’s correlation coefficient r2 = 0.993. Figure from Francioli et al. (2015) (Supplemental Figure 6).
An r-squared of 0.993 means that the processes we observe producing mutations in genomes explains 99.3% of the differences observed between the human and chimp genomes. You don't often see this tight of a regression in biology or genetics.
Transitions and transversions are the two main categories of substitution mutations, but there is yet another subset below that. These are CpG mutations. CpG is short for Cytosine-phosphate-Guanine. In other words, CpG is a CG in basic DNA sequence notation. For example, this randomly generated DNA sequence:
Are there any CpG's? Yes, there are 3 CpG's.
So what is so different about CpG's? In eukaryotes (like humans and other primates) the C in a CpG can be methylated. That is, a methyl group is covalently attached to the base. This can lead to a process called deamination which produces a T.
So this is a case of a mutation occurring in place instead of a base mismatch occurring during replication. CpG mutations occur at a much higher rate on a per base basis than do other types of mutations. The important thing to note is the use of rates in this example. There are far fewer CG in a sequence than there are single A, T, C, and G. If we look at the rate we are asking how often these types of mutations occur compared to where they could occur. So we aren't looking at the simple sum of all types of mutations, but their rates. In this context CpG transition mutations occur at a much, much higher rate than do other types mutations, and we see that in the multiple papers that directly measured mutation rates in human parent-offspring trios.
So let's go back to the Francioli paper and figure in message 4. They looked at the rate of mutations in all possible three base sequence. I created a list of all three base combinations in python and then highlighted all of the CpG's.
As you can see, there are 4 triplets that have CpG's. Therefore, there are four triplets that should see a much higher rate of C to T transitions than any other three base combination. So what do we see in the Francioli paper? Take a look at the figure in message 4. There are a group of 4 triangles all by themselves towards the upper end of the regression line. They are transitions given the color and the key in the figure (C:G>T:A) and they are CpG mutations indicated by the triangle shape of the data points. Those 4 data points off by themselves are the CpG transitions, and the rate of CpG mutations observed in humans is the same as the rate of CpG transitions seen between the comparison of the human and chimp genomes.
It can be seen even more clearly in a bar plot created by Stephen Schaffner:
Those sets of bar graphs did not include the de novo rates which is why I discussed the figure from the Francioli paper.
In combination, we can see that CpG are observed to occur at the highest rates in humans, CpG mutations occur at the highest rate when looking at standing human genetic variation, and (most importantly) CpG mutations occur at the highest rate when comparing the human and chimp genomes.
This is yet another massive and obvious fingerprint of mutation and common descent in the genetic data demonstrating that humans and chimps do share a common ancestor and that the differences between our genomes was produced by the very same processes we see operating in nature today.
There is no scientific evidence to support the wild conspiracy theory that all life came from one cell...
There is scientific evidence that all primates share a common ancestor, and that evidence was presented in this thread. Given your inability to actually address that evidence I can only assume it is really, really good evidence.